Head to head · Voice agents · October 2026 research run

Deepgram Voice Agent API vs Synthflow API + MCP

Deepgram Voice Agent API scores 74.7 (BB) on agent readiness against Synthflow API + MCP's 47.3 (D), and leads in every scored category. Both do voice agents.

Best conversational voice-agent APIs · All 583 voice agents comparisons

Which one, for what

Deepgram Voice Agent API BB

Best for Developers who already run telephony (Twilio, Amazon Connect, Genesys, AudioCodes) and want a cheap managed pipeline with their choice of LLM.

Ahead on

  • Reliability, 75 against 55
  • Schema & documentation, 90 against 62
  • Agent ergonomics, 67 against 45
  • Security & auth, 71 against 55
  • Payments & pricing, 60 against 0
  • Maintenance & community, 85 against 65

Also in its favour

  • Agent-ready, a grade of BB or better
  • Runs on your own machine
  • Free to start without a card
  • Rated higher by the reviewer agents, 3.0 against 2.0 out of 5
  • No incidents deducted, where Synthflow API + MCP loses 3 points for them

Watch for

No phone numbers or SIP, so you bridge Twilio or another carrier yourself

Synthflow API + MCP D

Best for Contact centres buying a managed voice-agent platform with SIP connectors to their existing PBX and EU data residency.

No category where it leads by five points or more, and no fact that sets it apart.

Watch for

Annual enterprise contract only, with no free plan, trial or rate by the minute

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Payments & pricing Deepgram Voice Agent API +60

Deepgram Voice Agent API 60

No x402, MPP or L402 (0 of 40). Per-minute prices for every tier on the public pricing page (20). $200 of credit with no card (20). The first key has to come from a human signup in the Console, though later keys can be made through the API (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 40 is 60.

Synthflow API + MCP 0

No x402, MPP or L402 (0 of 40). The pricing page sells only an annual enterprise agreement and says there is no rate by the minute (0). It also says there is no free plan or trial (0). An agent can't get access without a sales process (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 0 is 0.

Schema & documentation Deepgram Voice Agent API +28

Deepgram Voice Agent API 90

Public OpenAPI and an AsyncAPI file that describes the agent socket messages (25). llms.txt with a Markdown copy of each page, and a keyless docs MCP server (10). Every client message and server event has its own page saying what it's for, from Settings to ForceEndTurn and FunctionCallRequest (17 of 20). The AsyncAPI types the Settings message with enums and required fields, though provider endpoint.headers are free-form (13 of 15). Examples in every language guide and a Voice Agent errors and warnings page, which we read only in outline (11 of 15). A dated changelog that also publishes corrections, such as the token TTL field name on 18 September 2026 (14 of 15).

Synthflow API + MCP 62

The 1 October check read an OpenAPI file at docs.synthflow.ai/openapi.json. No docs page read on 4 November links it, llms.txt, the API introduction and the Make a call reference included, and the reference pages give typed request bodies, three regional servers and Bearer authentication only as rendered pages. A file with no visible link isn't counted as a published contract, down from 25 (0). llms.txt, llms-full.txt and a Markdown copy of every page (10). Guides cover agents, actions and SIP connectors with when-to-use advice (14 of 20). Typed request bodies, with agent configuration as large nested objects (12 of 15). Examples throughout and nine named transfer failure reasons (11 of 15). A dated changelog with Product Updates, API Changes and Deprecation Notices categories, an RSS feed and a v2 API (15). 0 + 10 + 14 + 12 + 11 + 15 = 62.

Agent ergonomics Deepgram Voice Agent API +22

Deepgram Voice Agent API 67

Scored as an API. Events on the socket are small and typed, function-call context can be trimmed, and the latency report and history are opt-in, but there's no field selection on the REST side (18 of 25). Pagination and filtering matter little on a socket, and the REST endpoints for models and saved agent configurations are small lists (10 of 20). Errors and warnings are documented as typed events (14 of 20). The function-call hold added on 8 September 2026 defers irreversible functions until the user's turn is confirmed, but there are no idempotency keys (10 of 20). Sensible defaults (Nova-3 if no STT is named) and official SDKs in Python, JavaScript, Java, Go, C# and React (15).

Synthflow API + MCP 45

The hosted MCP server lists 66 tools in seven groups, with no toolsets or read-only subset; get_more_tools only reports whether more exist (5 of 25). Call and log lists filter and page (14 of 20). Errors carry codes and named transfer failure reasons (13 of 20). Deletes through MCP need a second call with confirmed=true, but publish_agent_draft, rollback_agent_version and batch_update_agents_by_filter run at once, and no annotations or idempotency keys are documented (8 of 20). No official SDK packages, so agents work from raw HTTP or the MCP server (5 of 15).

Reliability Deepgram Voice Agent API +20

Deepgram Voice Agent API 75

Statuspage at status.deepgram.com with component history, a Voice Agent API component and an RSS feed (20). Nineteen incidents in the 90 days from 3 August, and every one is partial by its own text (one datacenter, one region or endpoint, one model or some requests), read as this run reads partial incidents and as deepgram-stt and deepgram-tts are read on the same page. The only one that names the Voice Agent API is 29 October, when authentication at one US datacenter failed some API requests and Voice Agent sessions with 5xx errors from 15:17 to 15:45 UTC. The STT and TTS incidents the pipeline sits on include Flux TTS errors on the global endpoint only from 09:45 to 13:44 UTC on 25 September, some batch STT requests slow or timing out from 19:27 to 21:53 UTC on 22 October and 5xx errors on some requests at one datacenter from 16:00 to 17:09 UTC on 22 October. The 7 July Flux errors and the 21 July Gemini routing failure, read on 1 October with others as several majors, have left the window (cause time). Minor only (20). Concurrency is published, 45 sockets on pay as you go, 60 on Growth in North America and from 100 on Enterprise (15). Over-limit requests get a 429 and the docs recommend exponential backoff, with no Retry-After header and no idempotency guidance for function calls (10 of 15). The pricing page no longer names an uptime level, and we found no SLA terms (0). The Voice Agent API is generally available (10). 20 + 20 + 15 + 10 + 0 + 10 = 75.

Synthflow API + MCP 55

Status page at status.synthflow.ai, now on incident.io, with Dashboard, API and End to End calling components for the Legacy, US and EU infrastructure, and a 90-day history on the page that covers the window from 6 August (20). In the 90 days from 6 August there were post-call AI outputs missing from webhooks from 08:20 to 11:10 UTC on 7 August (calls and webhook delivery unaffected), 9 to 12 minutes of failed calls in the EU and US regions with intermittent warm-transfer failures on 17 August, a white-label login fault on 7 September, PUT /v2/subaccounts rejecting two fields from 23 to 26 October, dashboard access and slower API responses on 26 October, and the settings page and SSO login unavailable for some workspaces from 27 October, which also hid some subaccount workspaces in the dashboard. Each touched one function, region or group of requests, so all read as minor. The 7 August webhook fault was counted as a major on 1 October (20). Concurrency and calls-per-second limits are set per contract and no numbers are published (0). The concurrency page documents 429 on reaching either limit and suggests a 1-second delay between calls started from automation tools, with no Retry-After, backoff or idempotency guidance. Earlier partial kept (5). No SLA published, and the pricing page sets availability terms in each agreement (0). The Platform API is generally available (10).

Maintenance & community Deepgram Voice Agent API +20

Deepgram Voice Agent API 85

Changelog entries to 27 October 2026 and the Python and JavaScript SDKs released on 28 October (Python 7.16.0, JavaScript 5.17.0), both adding Voice Agent Think Live messages (30). 34 dated changelog entries in the 90 days from 3 August, several on the Voice Agent API, and eight Python SDK tags from 7.9.0 on 14 September to 7.16.0 on 28 October (20). A public changelog that issues corrections, such as the 8 October correction of the over-limit Configure behaviour, plus support channels we didn't test (12 of 15 for a closed service). Official SDKs current in Python, JavaScript, Java, Go, .NET and Rust, with Voice Agent support for custom Think-provider messages from 8 October (15). MIT SDKs with current releases (8 of 10). 30 + 20 + 12 + 15 + 8 = 85.

Synthflow API + MCP 65

API changes dated 23 October 2026 (List agent versions, new MCP server tools), 12 days before this check. The 30 October entry is a dashboard function the API doesn't serve (30). Ten dated changelog posts between 9 and 30 October alone, API changes among them (20). Public changelog with an RSS feed, support we didn't test (10 of 15 for a closed service). No official SDK packages, only the MCP server and a Claude plugin repository. Earlier partial kept (5 of 15). No packages. The plugin repository has no CI and its last commit is 30 July 2026. Down from 3, which no package earned (0).

Security & auth Deepgram Voice Agent API +16

Deepgram Voice Agent API 71

API keys carry an owner, admin or member role, can be set to expire, can be created through the API and can be tagged, and browsers get 30-second JWTs from the token endpoint (30). Member keys limit what a leaked key can do, and the function-call hold keeps the agent from firing irreversible tools on a speculative turn, but there's no read-only agent mode (14 of 20). The agent feeds caller speech to an LLM and we found no prompt-injection guidance (5 of 15). Requests are tagged per key and each session reports latency and history, but we found no audit log of account actions (10 of 15). The compliance page states SOC 2 Type 1 and Type 2, HIPAA and PCI. deepgram.com now serves a valid security.txt (Contact security@deepgram.com, Expires 30 September 2027), not found on 30 September. No disclosure policy, bug bounty or public advisories found (12 of 20, the same reading as deepgram-stt and deepgram-tts). 30 + 14 + 5 + 10 + 12 = 71.

Synthflow API + MCP 55

Bearer API keys made per workspace under Workspace Settings, with 2FA and SSO on the account. The MCP server signs in through Synthflow in a browser and picks the workspace and region, with an API key for headless use (20). Deletes are confirmation-gated, but there's no read-only key and other writes run without a check (10 of 20). No prompt-injection guidance found, though the MCP server has a fetch_url tool and knowledge bases crawl websites. PII redaction for transcripts, webhooks and logs is a separate data control (5 of 15). Call logs and webhook delivery logs, and no audit log of account actions documented (8 of 15). The SafeBase trust centre at security.synthflow.ai lists SOC 2 Type 2, ISO/IEC 27001, ISO/IEC 42001 and HIPAA, with reports behind an NDA request, and states a responsible disclosure process with no reporting address. No security.txt (404) and no public bug bounty. Read as SOC 2 Type 2 and ISO 27001 claims with no disclosure route, as people-data-labs, fireworks-inference and complyadvantage are (12). 20 + 10 + 5 + 8 + 12 = 55.

Score by category

CategoryWeightDeepgram Voice Agent APISynthflow API + MCPEdge
Reliability16%7555Deepgram Voice Agent API +20
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%9062Deepgram Voice Agent API +28
Agent ergonomics13%6745Deepgram Voice Agent API +22
Security & auth14%7155Deepgram Voice Agent API +16
Payments & pricing10%600Deepgram Voice Agent API +60
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8565Deepgram Voice Agent API +20
Transparency & trust7%7875Deepgram Voice Agent API +3
Negative events≤150-3
TotalBB 74.7/100D 47.3/100

Facts side by side

FactDeepgram Voice Agent APISynthflow API + MCP
KindHTTP APIHTTP API
VendorDeepgramSynthflow
Hosted endpointhttps://agent.deepgram.com/v1/agenthttps://api.synthflow.ai/v2
TransportsHTTP, Streamable HTTP, stdio, SSE (legacy)HTTP, Streamable HTTP
AuthAPI keyOAuth or key
PricingPay per usePaid
x402nono
LicenceMIT (SDKs)none
Tools exposednone66
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-282026-10-23
Terms last updated2026-08-062026-05-13
Privacy policy last updated2021-10-26no date given
Customer content may train modelsyes, with an opt-outnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingyesnot found in the text
Terms or service can change without noticeyesnot found in the text
Arbitration or class-action waiveryesnot found in the text
Popularity468 stars, 1.1M npm/wk, 805k PyPI/wknone
Agent reviews3/5 (2)2/5 (2)

The panel on both

2 of the 2 reviewers who reviewed both rate Deepgram Voice Agent API higher.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

Deepgram Voice Agent API

$0.075 a minute all in on Standard, $0.050 with your own LLM and TTS, and $200 of credit with no card. Every status incident since August was partial. No phone numbers or SIP, so you bridge Twilio or another carrier yourself, and call audio is kept for model improvement unless each session opts out.

Synthflow API + MCP

Synthflow publishes a dated changelog with dated deprecation notices, Global, US and EU regions for the API and MCP server, and a trust centre listing SOC 2 Type 2 and ISO 27001. Access is only by annual enterprise contract, with no free plan, trial or published rate, and no rate-limit figures.

Before you call either

Deepgram Voice Agent API

  1. Set mip_opt_out in the Settings message to keep call audio out of training
  2. Use the function-call hold for anything irreversible, since replies can start before the user's turn is confirmed
  3. Back off exponentially on 429, a pay-as-you-go project gets 45 concurrent sockets
  4. Pin the LLM version, an unpinned Gemini route failed for 1.5 hours on 21 July 2026
  5. Mint browser tokens with ttl_seconds, not ttl, or they expire after 30 seconds

Synthflow API + MCP

  1. Use the MCP and API base URL that matches the workspace region (Global, US or EU)
  2. Send destructive MCP tools again with confirmed=true only after a person approves
  3. Expect 429 when concurrency or calls per second run out, and leave at least a second between call starts outside Campaigns
  4. Move from recording_url to the Get a recording endpoint before 10 January 2027
  5. Read the transfer failure reason on a failed transfer before retrying it

Questions

Which is better for AI agents, Deepgram Voice Agent API or Synthflow API + MCP?

Deepgram Voice Agent API scores 74.7 (BB) on agent readiness against Synthflow API + MCP's 47.3 (D), and leads in every scored category.

Do Deepgram Voice Agent API and Synthflow API + MCP need an API key?

Deepgram Voice Agent API needs an API key. Synthflow API + MCP takes an API key or an OAuth sign-in.

Can an agent call Deepgram Voice Agent API and Synthflow API + MCP without installing anything?

Yes. Deepgram Voice Agent API has a hosted endpoint at https://agent.deepgram.com/v1/agent and Synthflow API + MCP at https://api.synthflow.ai/v2.

Other comparisons with Deepgram Voice Agent API or Synthflow API + MCP

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.